Diese Seite dokumentiert Produktionsaktualisierungen für AutoML Natural Language. Prüfen Sie diese Seite regelmäßig auf Hinweise zu neuen oder aktualisierten Features, Fehlerkorrekturen, bekannten Problemen und verworfenen Funktionen.
Die neuesten Produktaktualisierungen für Google Cloud finden Sie auf der Seite mit den Google Cloud-Versionshinweisen.
Für neueste Produktaktualisierungen können Sie die URL der Seite in Ihren Feedreader einfügen oder die Feed-URL direkt hinzufügen:
April 03, 2020
Integration with VPC Service Controls is now in beta stage.
March 20, 2020
AutoML Natural Language now supports TIFF files as training data and input for predictions.
AutoML Natural Language now supports classification and sentiment analysis in 20 languages.
January 30, 2020
AutoML Natural Language now supports entity analysis in over 100 languages.
December 12, 2019
AutoML Natural Language General Availability (GA) release
AutoML Natural Language enables you to train three types of custom machine learning models:
- AutoML Text & Document Classification creates models to classify documents according to labels that you define.
- AutoML Text & Document Entity Extraction creates models to identify a custom set of entities within documents.
- AutoML Text & Document Sentiment Analysis creates models to analyze attitudes within text.
The GA release consolidates these features, which were documented as separate products during beta testing. See the AutoML Natural Language documentation for details about all features.
October 30, 2019
AutoML Natural Language offers you control over where your datasets and models are stored and where machine learning takes place. In particular, you can configure AutoML Natural Language to at rest and perform machine learning processing in the European Union only. See Locations for more information.
October 21, 2019
AutoML Natural Language Text Classification now supports batch prediction.
AutoML Natural Language Text Classification can now classify PDF files and also supports PDF files as training items.
September 04, 2019
AutoML Natural Language Entity Extraction offers support for document-based entity extraction, which enables you to identify entities in a format that recreates the basic layout of the training item. When you add annotations, AutoML Natural Language Entity Extraction captures the annotation's position on the page as a factor considered during training. See Managing datasets for more information.
April 10, 2019
AutoML Natural Language Entity Extraction Beta has been released.
AutoML Natural Language Sentiment Analysis Beta has been released.
Microsoft Edge and Microsoft Internet Explorer do not support all features of AutoML Natural Language Entity Extraction. If you are having problems, try Google Chrome, Safari, or Firefox.
Microsoft Edge and Microsoft Internet Explorer do not support all features of AutoML Natural Language Sentiment Analysis. If you are having problems, try Google Chrome, Safari, or Firefox.
January 22, 2019
Beginning January 22, 2019, inactive models are subject to automatic undeployment. An inactive model is one that has not been used for prediction in 60 days. An undeployed model is not available for use until you explicitly redeploy it using a method that will be made available in advance of any undeployments.
November 01, 2018
The AutoML Natural Language Text Classification service account requires additional permissions to access project resources when using the AutoML Natural Language Text Classification UI. You must add the role
serviceusage.serviceUsageAdmin. See step 8 in Before You Begin.
August 24, 2018
The URL for the AutoML Natural Language Text Classification UI has changed to
https://console.cloud.google.com/natural-language. The previous URL
https://beta-dot-custom-vision.appspot.com/text/ will remain active.
July 24, 2018
AutoML Natural Language Text Classification Beta Release
Models created during the EAP will continue to be valid for the Beta. You do not need to recreate/retrain models. However, the format of resource IDs (dataset IDs, model IDs, and operation IDs) is changing. If you are using hardcoded resource IDs, you need to obtain new IDs for those resources (using the get or list method) by August 31, 2018.
Note: Enabling the AutoML API for your project automatically also enables the Cloud Vision API, Cloud Translation API, and Cloud Natural Language API. Disabling any of these three APIs will also disable the AutoML API.
June 07, 2018
The earlier version of the AutoML Natural Language Text Classification UI has been deprecated. Attempts to connect to the older UI are redirected to the new UI.
May 21, 2018
The refresh of the AutoML Natural Language Text Classification EAP release introduces a new UI design and a new API.
April 02, 2018
AutoML Natural Language Text Classification has changes to its modeling logic. You may notice an increase in training latency as well as potential improvements in training performance.
February 07, 2018
AutoML Natural Language Text Classification EAP release